Laboratory bench tests with improved ability to predict grease lubrication performance in rolling bearings
File(s)
Author(s)
McAllister, Rory
Type
Thesis
Abstract
Frictional losses in rolling bearings produce an enormous quantity of wasted energy due to the abundance of bearings in the modern world. This loss can be reduced by improving the frictional performance of the lubricating greases that are the primary lubricant of bearings. A vital part of this improvement, and of developing new greases, is the test system used. It can provide deep insight into the observed phenomena, or potentially mislead, if the relevant lubrication mechanisms are not accurately emulated.
Bearing tests were conducted under conditions representative of typical applications, with a set of greases containing some of the most common thickener types. The bearings were dismantled post-test and imaged, and the lubricant films were analysed in-situ using IR spectroscopy to assess their chemical composition. The bearings showed a typical grease distribution, with the majority situated at the raceway shoulders and shields and only a small quantity of ‘active’ lubricant remaining in the raceway. Certain IR changes were associated with shear degradation, and these were used in a novel way to assess the extent of degradation. The active lubricant was found to be oil-rich and heavily shear-degraded.
Rheometry and conventional ball-on-disc testing of friction and film thickness were applied to the fresh test greases, as well as shear degraded and thermally aged samples. The degraded samples behaved very differently to fresh grease, suggesting that testing of fresh grease does not represent the friction, film thickness and rheology of the active lubricant observed in bearings. Degradation during ball-on-disc testing was compared to that of bearing tests and was found to be far more limited. Inherent problems with the ball-on-disc geometry were also identified that limit applicability to the contacts of bearings.
From these findings, a novel barrel-on-ring test device was developed which aimed to replicate bearing conditions in the most salient ways. Comparison of film thickness in the new device, a ball-on-disc machine and bearings under the same starvation conditions showed that the new rig allowed reflow of grease at a much greater rate than the ball-on-disc device, approaching the film thickness of the bearings. This suggests that the novel device is more representative of bearing lubrication than standard tests, and therefore may provide better predictions about bearing performance, as well as better understanding of lubrication phenomena.
Bearing tests were conducted under conditions representative of typical applications, with a set of greases containing some of the most common thickener types. The bearings were dismantled post-test and imaged, and the lubricant films were analysed in-situ using IR spectroscopy to assess their chemical composition. The bearings showed a typical grease distribution, with the majority situated at the raceway shoulders and shields and only a small quantity of ‘active’ lubricant remaining in the raceway. Certain IR changes were associated with shear degradation, and these were used in a novel way to assess the extent of degradation. The active lubricant was found to be oil-rich and heavily shear-degraded.
Rheometry and conventional ball-on-disc testing of friction and film thickness were applied to the fresh test greases, as well as shear degraded and thermally aged samples. The degraded samples behaved very differently to fresh grease, suggesting that testing of fresh grease does not represent the friction, film thickness and rheology of the active lubricant observed in bearings. Degradation during ball-on-disc testing was compared to that of bearing tests and was found to be far more limited. Inherent problems with the ball-on-disc geometry were also identified that limit applicability to the contacts of bearings.
From these findings, a novel barrel-on-ring test device was developed which aimed to replicate bearing conditions in the most salient ways. Comparison of film thickness in the new device, a ball-on-disc machine and bearings under the same starvation conditions showed that the new rig allowed reflow of grease at a much greater rate than the ball-on-disc device, approaching the film thickness of the bearings. This suggests that the novel device is more representative of bearing lubrication than standard tests, and therefore may provide better predictions about bearing performance, as well as better understanding of lubrication phenomena.
Version
Open Access
Date Issued
2022-09
Date Awarded
2023-04
Copyright Statement
Creative Commons Attribution NonCommercial Licence
License URL
Advisor
Cann, Philippa
Masen, Marc
Sponsor
Shell
Grant Number
16000154
Publisher Department
Mechanical Engineering
Publisher Institution
Imperial College London
Qualification Level
Doctoral
Qualification Name
Doctor of Philosophy (PhD)